Building and training a fuzzy neural model of data mining tasks

Abstract

Abstract In this paper discussed the problem of constructing and training a fuzzy neural network based on fuzzy logical rules. Based on the constructed model, the developed algorithm, the objects classified with indistinct initial information. Under these conditions, traditional methods of mathematical statistics or simulation modeling do not allow building adequate models for solving data mining problems. We consider the solution of problems based on the use of intelligent modeling technologies using methods of neural networks and fuzzy logic in order to automate the construction of fuzzy rules based on methods and algorithms of machine learning.